Attributes and predictors of long COVID

Area of Science:

• Epidemiology
• Infectious Diseases
• Public Health

Background:

• Rising reports of long-lasting coronavirus disease 2019 (COVID-19) symptoms, termed 'long COVID', highlight knowledge gaps in prevalence and predictive factors.
• Understanding the trajectory of COVID-19 symptoms is crucial for managing patient outcomes and healthcare resource allocation.

Purpose of the Study:

• To determine the prevalence of long COVID symptoms.
• To identify risk factors associated with developing long COVID.
• To develop a predictive model for identifying individuals at risk of a protracted COVID-19 course early in the disease.

Main Methods:

• Analysis of prospective symptom data from 4,182 individuals using the COVID Symptom Study app.
• Categorization of symptom duration into periods of ≥28 days, ≥8 weeks, and ≥12 weeks.
• Development and validation of a predictive model using early symptom data (7 days post-infection) to distinguish between short and long COVID.

Main Results:

• 13.3% of participants reported symptoms lasting ≥28 days, with 4.5% experiencing symptoms for ≥8 weeks.
• Long COVID was associated with increased age, body mass index, and female sex.
• Experiencing >5 symptoms in the first week significantly increased the odds of long COVID (OR=3.53).
• A predictive model developed at 7 days showed 76% accuracy in distinguishing long COVID, validated in an independent cohort.

Conclusions:

• A significant proportion of individuals experience prolonged symptoms after COVID-19 infection.
• Early symptom burden is a key predictor of long COVID development.
• The developed predictive model offers a tool for early identification of at-risk individuals for targeted interventions and resource planning.